README.md

Data mining project

Project developed for the undergraduate elective course "Data mining and Machine Learning" at CEID.

Part A

Dataset: winequality-red.csv

• Suport Vector Machines (SVM)
• Missing data handling
  1. Drop column
  2. Fill NaN values with column average
  3. Logistic Regression imputation
  4. Imputation based on K-means

Evaluation metrics: f1 score, precision, recall and accuracy

Part B

Dataset: onion-or-not.csv

• Data preprocessing (NLTK)
  1. Word tokenizer
  2. Stemming
  3. Stopwords removal
  4. Tf-idf matrix
• Neural network (Tensorflow keras)

Evaluation metrics: f1 score, precision, recall and accuracy

Authors

• Zisis Stylianos Tramparis
• Romanos Kapsalis